Investigating the usefulness of satellite derived fluorescence data in inferring gross primary productivity within the carbon cycle data assimilation system

(1)

BGD

12, 707–749, 2015

Investigating the usefulness of satellite derived fluorescence data

E. N. Koffiet al.

Title Page

Abstract Introduction

Conclusions References

Tables Figures

◭ ◮

Back Close

Full Screen / Esc

Printer-friendly Version Interactive Discussion

Discussion

P

a

per

|

Discussion

P

a

per

|

Discussion

P

a

per

|

Discussion

P

a

per

|

Biogeosciences Discuss., 12, 707–749, 2015 www.biogeosciences-discuss.net/12/707/2015/ doi:10.5194/bgd-12-707-2015

This discussion paper is/has been under review for the journal Biogeosciences (BG). Please refer to the corresponding final paper in BG if available.

Investigating the usefulness of satellite

derived fluorescence data in inferring

gross primary productivity within the

carbon cycle data assimilation system

E. N. Ko_ffi1,*, P. J. Rayner2, A. J. Norton2, C. Frankenberg3, and M. Scholze4

1

Laboratoire des Sciences du Climat et de l’Environnement (LSCE), UMR8212, Ormes des merisiers, 91191 Gif-sur-Yvette, France

2

School of Earth Sciences, University of Melbourne, Melbourne, Australia

3

Jet Propulsion Laboratory, California Institute of Technology, Pasadena, USA

4

Department of Physical Geography and Ecosystem Science, Lund University, Lund, Sweden

*

now at: the European Commission Joint Research Centre, Institute for Environment and Sustainability, 21027 Ispra (Va), Italy

Received: 1 December 2014 – Accepted: 15 December 2014 – Published: 13 January 2015

Correspondence to: E. N. Koffi(ernest.koffi@jrc.ec.europa.eu)

(2)

BGD

12, 707–749, 2015

E. N. Koffiet al.

Title Page

Tables Figures

◭ ◮

Back Close

Full Screen / Esc

Discussion

P

a

per

|

Discussion

P

a

per

|

Discussion

P

a

per

|

Discussion

P

a

per

|

Abstract

We investigate the utility of satellite measurements of chlorophyll fluorescence (Fs) in

constraining gross primary productivity (GPP). We ingest Fs measurements into the

Carbon-Cycle Data Assimilation System (CCDAS) which has been augmented by the fluorescence component of the Soil Canopy Observation, Photochemistry and Energy 5

fluxes (SCOPE) model. CCDAS simulates well the patterns ofFssuggesting the

com-bined model is capable of ingesting these measurements. However simulated Fs is

insensitive to the key parameter controlling GPP, the carboxylation capacity (Vcmax).

SimulatedFs is sensitive to both the incoming absorbed photosynthetically active

radi-ation (aPAR) and leaf chlorophyll concentrradi-ation both of which are treated as perfectly 10

known in previous CCDAS versions. Proper use ofFsmeasurements therefore requires

enhancement of CCDAS to include and expose these variables.

1 Introduction

The natural terrestrial flux has been identified as the most uncertain term in the global carbon budget (Le Quere et al., 2013). The gross primary productivity (GPP), which is 15

the flux of CO₂assimilated by plants during photosynthesis, is the input to this system so its variation can significantly contribute to the uncertainties in terrestrial CO2fluxes.

Complex systems have been built to reduce the uncertainties in GPP. These systems are either based on up-scaling or atmospheric inverse modeling methods. Up-scaling methods estimate GPP at global scale by establishing relationships between local GPP 20

measurements and environmental variables then using these variables to calculate GPP globally (e.g., Jung et al., 2011; Beer et al., 2010 and references therein). The inverse modeling approach uses CO2 concentration observations at global scale to

constrain the process parameters of carbon models that compute the terrestrial fluxes. This inverse method is an example of Carbon Cycle Data Assimilation Systems (CC-25

DAS). The CCDAS considered in the present study has two main components:

(3)

BGD

12, 707–749, 2015

E. N. Koffiet al.

Title Page

Tables Figures

◭ ◮

Back Close

Full Screen / Esc

Discussion

P

a

per

|

Discussion

P

a

per

|

Discussion

P

a

per

|

Discussion

P

a

per

|

– A deterministic dynamical model that computes the evolution of both the bio-sphere and soil carbon stores given an initial condition, forcing and a set of the model process parameters.

– An assimilation system that allows the adjustment of a subset of the state vari-ables, initial conditions and/or process parameters to reduce the mismatch be-5

tween the model simulations and observations. Usually any prior information on the variables which are adjusted are also taken into account (see e.g., Kaminski et al., 2002, 2003; Rayner et al., 2005, and references therein for the underly-ing methodology). In the above references the target variables are the process parameters of the Biosphere Energy-Transfer Hydrology (BETHY) model (Knorr, 10

2000).

Rayner et al. (2005) built such a system around the biosphere model BETHY coupled to atmospheric transport models, see also Kaminski et al. (2013) for an overview on fur-ther developments and applications. Koffi_{et al. (2012) used this CCDAS to investigate} the sensitivity of estimates of GPP to transport models and observational networks 15

of CO2 concentrations. Large differences in GPP in the tropics were found between

their estimates and those from either satellite based products or up-scaling methods. Koffi_{et al. (2012) found significantly larger GPP in the tropics where the parameters of} BETHY are weakly constrained due to few CO2 concentration observations available

in this region. 20

Recent work have inferred plant fluorescence (hereafter Fs) from the Greenhouse

gas Observing Satellite (GOSAT; e.g., Frankenberg et al., 2011, 2012; Joiner et al., 2011; Guanter et al., 2012), ENVISAT/SCIAMACHY (Joiner et al., 2012), and MetOp-A/GOME-2 (Joiner et al., 2013). They showed thatFs data are promising for inferring

GPP. They found a strong linear correlation between satellite-basedFs and GPP

esti-25

(4)

BGD

12, 707–749, 2015

E. N. Koffiet al.

Title Page

Tables Figures

◭ ◮

Back Close

Full Screen / Esc

Discussion

P

a

per

|

Discussion

P

a

per

|

Discussion

P

a

per

|

Discussion

P

a

per

|

zones where estimates from a CCDAS are found to be uncertain. It is worth asking whether such fluorescence data is useful to constrain GPP in the CCDAS framework.

The relationship between fluorescence and photochemistry at leaf level is reason-ably well understood. Light energy absorbed by chlorophyll molecules has one of three fates: photosynthesis, dissipation as heat (non-photochemical quenching) or chloro-5

phyll fluorescence. The total amount of chlorophyll fluorescence is only 1 to 2 % of total light absorbed. The spectrum of fluorescence is diff_{erent to that of absorbed light. The} peak of the fluorescence spectrum lies between 650 and 850 nm. Under low light condi-tions, a negative correlation has been found between fluorescence and photosynthesis light use effi_{ciencies (e.g., Genty et al., 1989; Rosema et al., 1998; Seaton and Walker,} 10

1990; Maxwell and Johnson, 2000; van der Tol et al., 2009). At high light conditions (i.e., high irradiance and moisture stress), a positive correlation has been observed be-tween fluorescence and photosynthesis light use effi_{ciencies (Gilmore and Yamamoto,} 1992; Gilmore et al., 1994; Maxwell and Johnson, 2000; Van der Tol et al., 2009). Re-garding the water stress, more recently, Jung-See Lee et al. (2012) showed a negative 15

correlation between vapour pressure deficit andFs.

The cited works show that the link between fluorescence and photosynthesis is com-plex. Thus, before using fluorescence observations to constrain gross primary produc-tivity in the framework of CCDAS, we need first to ensure that there is a common parameter or set of parameters relevant to both the fluorescence and photosynthesis 20

process models of the CCDAS. So, if there are common parameters, we can assess the sensitivities of GPP andFsto them. This requires implementing in CCDAS a model

that allows computing both fluorescence and photosynthesis. We build such a CCDAS by using the Soil Canopy Observation, Photochemistry and Energy fluxes (SCOPE) model (Van der Tol et al., 2009a, 2014). SCOPE is based on the existing theory of 25

chlorophyll fluorescence and photosynthesis. The photosynthesis scheme of C3 plants uses the formulations of Collatz et al. (1991), while for the C4 photosynthesis pathway, the formulations of Collatz et al. (1992) are considered. In these formulations of the photosynthesis, the maximum carboxylation rateVcmaxis a key process parameter. The

(5)

BGD

12, 707–749, 2015

E. N. Koffiet al.

Title Page

Tables Figures

◭ ◮

Back Close

Full Screen / Esc

Discussion

P

a

per

|

Discussion

P

a

per

|

Discussion

P

a

per

|

Discussion

P

a

per

|

fluorescence model is based on the work of Genty et al. (1989), Rosema et al. (1998), and van der Tol et al. (2014). The model is formulated such that the sum of the prob-abilities of an absorbed photon to result in fluorescence, photochemistry, and heat is unity. Hence, the fluorescence model also utilizesVcmaxas a process parameter.

CCDAS operates in two modes (Scholze et al., 2007). The calibration mode that 5

derives an optimal parameter set including posterior uncertainties of the dynamical carbon model (here the biosphere model) by constraining the process parameters of the model with observations. The diagnostic/prognostic (referred hereafter as forward) mode allows deriving the various quantities of interest (e.g., terrestrial carbon fluxes or atmospheric CO₂ concentrations) and their uncertainties. These quantities are calcu-10

lated from the optimized parameter vector obtained from the calibration step. CCDAS has been widely applied to investigate terrestrial carbon cycling (e.g., Rayner et al., 2005; Scholze et al., 2007) and in particular more recently to (i) estimate the GPP at global scale (Koffi _{et al., 2012) and (ii) to quantify the uncertainty in the parameters} of BETHY by using both CO2concentration and flux observational networks (Kaminski

15

et al., 2012; Koffi_{et al., 2013). To assess the usefulness of satellite based fluorescence} data (Fs) to constrain GPP within CCDAS, we first build the forward mode of the

CC-DAS around the model SCOPE, which is used to investigate the sensitivities of both GPP andFsto the biochemical parameters as well as environmental conditions.

The work is organized as follows: 20

In Sect. 2, we describe both the model SCOPE and its coupling with CCDAS and the fluorescence data retrieved from the satellite GOSAT. In Sect. 3, we perform various idealized sensitivity tests to investigate the strength of the relationships betweenFsand

GPP by using the SCOPE model alone. These tests are performed by studying the sensitivity of GPP andFsto the biochemical parameters (i.e.,Vcmaxand the chlorophyll

25

content Cab) and the environmental conditions (e.g., short wave radiation Rin). The

vegetation is characterized by diff_{erent values of the leaf area index (LAI). In Sect. 4,} by using the forward mode of the CCDAS coupled to SCOPE, we compute bothFsand

(6)

BGD

12, 707–749, 2015

E. N. Koffiet al.

Title Page

Tables Figures

◭ ◮

Back Close

Full Screen / Esc

Discussion

P

a

per

|

Discussion

P

a

per

|

Discussion

P

a

per

|

Discussion

P

a

per

|

December 2010. The simulations are based on the diff_{erent settings of LAI,}_R_in_,_V_cmax_, andCabvalues. In Sect. 5, results are discussed. Finally, conclusions are presented in

Sect. 6.

2 Models and data

2.1 Models

5

2.1.1 SCOPE model

The model SCOPE is a 1-D model based on radiative transfer, micrometeorology, and plant physiology (van der Tol et al., 2009a). Version 1.53 of SCOPE is used in this study with the default version of the biochemical code (referred as fluorescence model choice “0”; van der Tol et al., 2014). SCOPE treats canopy radiative transfer in the visible and 10

infrared and chlorophyll fluorescence, as well as the energy balance. The modules of SCOPE are executed in the following order:

1. A semi-empirical radiative transfer model for incident sun and sky radiation, based on the SAIL model (Verhoef and Bach, 2007). This module calculates the outgoing radiation spectrum (0.4 to 50 µm) at the top of the canopy (hereafter TOC), as well 15

as the net radiation and absorbed photosynthetically active radiation (aPAR) per surface element.

2. A numerical radiative transfer model for thermal radiation generated internally by soil and vegetation, based on Verhoef et al. (2007). This module computes the TOC outgoing thermal radiation and net radiation per surface element, but for 20

heterogeneous leaf and soil temperatures.

3. A biochemistry model for C3 and C4 plants, which allows the computation of quan-tities relevant for photosynthesis and chlorophyll fluorescence at leaf level. At leaf level, the model calculates a fluorescence scaling factor relative to that of a leaf

(7)

BGD

12, 707–749, 2015

E. N. Koffiet al.

Title Page

Tables Figures

◭ ◮

Back Close

Full Screen / Esc

Discussion

P

a

per

|

Discussion

P

a

per

|

Discussion

P

a

per

|

Discussion

P

a

per

|

in low-light, unstressed conditions from absorbed fluxes, canopy and ambient en-vironmental conditions (radiation, temperature, water vapour, CO2, and O2

con-centrations).

4. A radiative transfer model for chlorophyll fluorescence based on the FluorSAIL model (Miller et al., 2005) that calculates the TOC radiance spectrum of fluores-5

cence over 640–850 nm from the geometry of the canopy and a calculated fluo-rescence spectrum that is linearly scaled by the leaf level chlorophyll fluofluo-rescence scaling factor.

SCOPE uses a canopy structure characterized by a spherical leaf angle distribution as a function of LAI with 60 distributed elementary layers. The geometry of the vegetation 10

is treated stochastically. SCOPE calculates the illumination of leaves with respect to their position and orientation in the canopy. The spectra of reflected and emitted radia-tion as observed above the canopy in the satellite observaradia-tion direcradia-tion are computed. It is worth noting that SCOPE permits variation only in the vertical dimension. Thus, it is valid for vegetation in which variations in the horizontal are smaller than in the ver-15

tical dimension. This is maybe a limitation for some natural canopies, especially when coupling to the CCDAS as performed in Sect. 2.1.2. However, the sensitivity of this limitation to the CCDAS results is beyond the scope of this study.

We briefly describe the fluorescence model at leaf level (more detail is given in van der Tol et al., 2009b and van der Tol et al., 2014) with focus on the variables and pa-20

rameters relevant for the photosynthesis. The model of Faquahar et al. (1980) divides photosynthesis into two main processes: (1) regeneration of the ribulose bisphosphate (RuP2), which depends on the light and (2) the maximum carboxylation rate at RuP2 saturated conditions in the presence of suffi_{cient light. The regeneration of RuP2 for} two photosystems (PSII and PSI) gives the link between photosynthesis and fluores-25

cence.

(8)

pho-BGD

12, 707–749, 2015

E. N. Koffiet al.

Title Page

Tables Figures

◭ ◮

Back Close

Full Screen / Esc

Discussion

P

a

per

|

Discussion

P

a

per

|

Discussion

P

a

per

|

Discussion

P

a

per

|

tochemistry, and heat is unity. Following this, the fluorescenceΦ_Ft_{from a single leaf is} calculated over the spectrum window of 640–850 nm as follows:

Φ_Ft= Φ_Fm₍₁₋Φ_p₎ ₍₁₎

WhereΦ_Fm _{is the fluorescence yield and computed as follows:}

Φ_Fm= Kf (Kf+Kd+Kn)

(2) 5

With

Kn=(6.2473x−0.5944)x (3)

Wherex stands for the degree of light saturation and defined as:

x=₁₋ Φ_p

Φ_p0 (4)

Φ_p_and Φ_p0 _{(given by the following expressions) stand for the fractions of actual and} 10

dark photochemistry yields, respectively:

Φ_p0= Kp (Kf+Kd+Kp)

(5)

Kf is the rate constant for fluorescence and sets to 0.05.

Kpis the rate constant for photochemistry with a value of 4.0.

Kd, with a value of 0.95, is the rate constant for thermal deactivation atΦFm.

15

Φ_p= Φ_p0Ja

Je

(6)

(9)

BGD

12, 707–749, 2015

E. N. Koffiet al.

Title Page

Tables Figures

◭ ◮

Back Close

Full Screen / Esc

Discussion

P

a

per

|

Discussion

P

a

per

|

Discussion

P

a

per

|

Discussion

P

a

per

|

Ja and Je stand for the actual and potential electron transport rates, respectively. Ja

is the electron transport rate used for gross primary productivity (GPP). van der Tol et al. (2014) used Pulse-Amplitude fluorescence measurements to derive an empirical relation between the effi_{ciencies of photochemistry and fluorescence. This relationship} was derived after analysing the response of non-photochemical quenching (NPQ) in 5

plants to light saturation. The formulations of GPP in SCOPE follow that of Collatz et al. (1991) and Collatz et al. (1992) for C3 and C4 plants, respectively. The poten-tial electron transport rateJe is related to the rate of absorbed photons (or absorbed

photosynthetically active radiation, i.e., aPAR), hence to the visible radiation. The flu-orescence is linearly related to the short wave (visible) radiation, while it is related to 10

Vcmax mainly when the gross primary productivity GPP is limited by the carboxylation

enzyme Rubisco and the capacity for the export or the utilization of the products of photosynthesis.

The total top-of-canopy fluorescent radiance is obtained by a summation of the flu-orescenceΦ_Ft _{(Eq. 1) from each of the leaves over all layers and orientations, taking} 15

into account the probabilities of viewing sunlit and shaded components. The model then calculates radiation transport in a multilayer canopy as a function of the solar zenith angle and leaf orientation to simulate fluorescence in the direction of satellite observation (Van der Tol et al., 2009a).

Leaf biochemistry aff_{ects reflectance, transmittance, transpiration, photosynthesis,} 20

stomatal resistance, and chlorophyll fluorescence. Reflectance and transmittance coef-ficients, which are a function ofCabare calculated by following the PROSPECT model

(Jacquemoud and Baret, 1990). Two excitation fluorescence matrices (EF-matrices) representing fluorescence from both sides of the leaf are computed. The matrices con-vert a spectrum of aPAR into a spectrum of fluorescence. Details on the radiative trans-25

(10)

BGD

12, 707–749, 2015

E. N. Koffiet al.

Title Page

Tables Figures

◭ ◮

Back Close

Full Screen / Esc

Discussion

P

a

per

|

Discussion

P

a

per

|

Discussion

P

a

per

|

Discussion

P

a

per

|

2.1.2 Coupling SCOPE to CCDAS

Within CCDAS we replace the radiative transfer and photosynthesis schemes of BETHY with their corresponding schemes from SCOPE and add the fluorescence model of SCOPE. The spatial resolution, vegetation characteristics as well as the me-teorological and phenological data of BETHY are used to force SCOPE. The spatial 5

resolution is 2◦

×2◦ with 3462 land grid points for the globe. CCDAS uses 13 plant

functional types (PFT) based on Wilson and Henderson-Sellers (1985). A grid cell can contain up to three diff_{erent PFTs, with the amount specified by their fractional} cover-age.

2.2 Data

10

2.2.1 GOSAT fluorescence data

Frankenberg et al. (2011, 2012), Joiner et al. (2011), and Guanter et al. (2012) have published maps ofFsfrom GOSAT (Kuze et al., 2009). The retrieval measures

terres-trial emission at the frequencies of solar Fraunhofer lines (gaps in the solar spectrum). Chlorophyll fluorescence is the main contributor to emissions at these frequencies. 15

GOSAT carries a Fourier Transform Spectrometer (FTS) measuring with high spec-tral resolution in the 755–775 nm range, which allows resolving individual Fraunhofer lines overlapping the fluorescence emission. The method described in Frankenberg et al. (2011) makes use of two spectral windows centered at 755 and 770 nm to derive

Fs. Results from the line centered around 755 nm for the period June 2009 to

De-20

cember 2010 are used in this study. The fluorescence data we are using are monthly means mapped onto 2◦

×2◦spatial resolution at global scale. The fluorescence product

includes uncertainties.

(11)

BGD

12, 707–749, 2015

E. N. Koffiet al.

Title Page

Tables Figures

◭ ◮

Back Close

Full Screen / Esc

Discussion

P

a

per

|

Discussion

P

a

per

|

Discussion

P

a

per

|

Discussion

P

a

per

|

2.2.2 Data relevant for models

The input data for the models we are using are of four main kinds: (i) the data for the radiative transfer modules of SCOPE, (ii) the data characterizing the environmental conditions (i.e., meteorological and short and long wave radiation) relevant for both the radiative transfer and biochemistry models, (iii) the leaf area index (LAI) for the radiative 5

transfer and biochemistry models, and (iv) the process parameters of the biochemistry models.

The model SCOPE requires incident radiation at the top-of-canopy as input. To take into account the atmospheric absorption bands properly, this data is needed at high resolution. The spectra of sun and sky fluxes at the top of the canopy are obtained from 10

the atmospheric radiative transfer model MODTRAN (Berk et al., 2000). MODTRAN was run for 16 atmospheric situations representative of diff_{erent regions (Verhoef et al.,} 2014). We use 4 types of these generated atmospheres. They are tropical atmosphere for the tropical zones, winter and summer atmospheres for high and middle latitudes. In addition, we have at our disposal data for an atmosphere which is representative of 15

the whole globe (hereafter “standard atmosphere”). We have tested the sensitivity of

Fsand GPP to these four types of atmospheres. Results show only residual differences

between the inferredFsand GPP. We consider the standard atmosphere for the

ideal-ized tests (Sect. 4.1) and the seasonal atmosphere for the simulations at global scale by using the CCDAS (Sect. 4.2).

20

The system needs forcing data to drive SCOPE within the CCDAS framework. Monthly observed climate, incident radiation, and fractional soil moisture for the pe-riod 2009–2010 are used (Weedon et al., 2011). The LAIs are obtained from BETHY simulation.

The main parameters that aff_{ect both the photosynthesis and fluorescence schemes} 25

are given in Table 1. The parameters are of two kinds: parameters that are PFT-specific (e.g.,Vcmax and Cab) and global parameters. Prior and optimized values of Vcmax

(12)

BGD

12, 707–749, 2015

E. N. Koffiet al.

Title Page

Tables Figures

◭ ◮

Back Close

Full Screen / Esc

Discussion

P

a

per

|

Discussion

P

a

per

|

Discussion

P

a

per

|

Discussion

P

a

per

|

nitrogen content of the leaf which itself is linked to the maximum rate of carboxyla-tion through the proteins of the Calvin Cycle and the thylakoids. Some investigators have related the photosynthetic capacity of leaves of some specific plants to their ni-trogen content (e.g., Evans, 1989; Kattge et al., 2009; Houborg et al., 2013). Other workers have derived some empirical relationships between the nitrogen content and 5

the chlorophyll content (e.g., Shaahan et al., 1999; Van den Berg and Perkins, 2004; Ghasemi et al., 2011). Since the current version of the model SCOPE does not include the nitrogen scheme of a leaf, we first use the same value of chlorophyll contentCab

for all 13 PFTs. As a second step,Cabvalues for each of the 13 PFTs are optimized so

that the simulatedF_sreproduces the main spatial characteristics of observedF_s. 10

3 Experimental set ups

3.1 Idealized tests

We carry out some idealized sensitivity tests by using the SCOPE model alone. We investigate the sensitivity of Fs and GPP to biochemical parameters Vcmax and Cab,

environmental variables (temperature, vapour pressure, etc), visible radiation, and LAI. 15

We assume throughout the following sections the concentrations of both CO2 and O2

at the interface of the canopy to be constant. We will focus our discussions on the assessment of the sensitivity of the simulatedFsand GPP toVcmax,Cab, and the short

wave radiation.

We present a spectrum of simulated fluorescence for C3 and C4 plants in Fig. 1. 20

Two peaks in the simulated fluorescence spectrum are shown at 680 and 725 nm. In agreement with van der Tol et al. (2009a), C4 plants exhibit largerFsthan C3 plants over

the wavelength range 625 to 755 nm. These diff_{erences are amplified around the two} peaks. We are using as observations the GOSAT satellite derivedF_s, which retrieved

Fs around 755 nm. Therefore, the simulated fluorescence in this study corresponds to

25

theFs value at this wavelength. In Fig. 1, this is around 1.2 W m− 2

µm−1_sr−1_.

(13)

BGD

12, 707–749, 2015

E. N. Koffiet al.

Title Page

Tables Figures

◭ ◮

Back Close

Full Screen / Esc

Discussion

P

a

per

|

Discussion

P

a

per

|

Discussion

P

a

per

|

Discussion

P

a

per

|

For all the idealized tests presented hereafter, we use 8 values of LAI: 0.1, 0.5, 1, 2, 3, 4, 5, and 6. We select these values to be able to characterize diff_{erent types of} canopy from sparse to dense vegetation. Also, the pressure, the temperature, and the water vapour pressure are set to 1000 hPa, 25◦_{C, and 10 hPa, respectively. The carbon} dioxide (CO2) and the oxygen (O2) concentrations are set to 355 ppm and 210×10

3

5

ppm, respectively. We consider the value of the simulated fluorescenceFsfrom SCOPE

at 755 nm.

– To investigate the sensitivity ofFsand GPP to the maximum carboxylation

capac-ityVcmax, we chooseVcmaxvalues ranging from 10 to 250 µmol (CO2) m− 2

s−1_every 10 µmol m−2_s−1_{. In addition, two small}_V

cmaxvalues of 0.5 and 5 µmol m− 2

s−1_are 10

considered.

– To study the sensitivity of Fs and GPP to the chlorophyll content AB (Cab) we

selectC_ab values that span 10 to 80 µg cm−2_{range every 5 µg cm}−2_{. Additionally,} a smallCabvalue of 1 µg cm−

2

is considered.

– To assess the sensitivity of theFsand GPP to the short wave radiation (Rin) at the

15

top of the canopy, we selectR_in values that range from 100 W m−2_{to 1300 W m}−2 every 100 W m−2_{. We add small values of 1, 5, 10, 25, 50, and 75 W m}−2_.

– Finally, to investigate the diurnal variations, we simulateFsand GPP by using the

short time series of half hourly data over 16–20 June 2006 over a canopy located at 52.25 deg. latitude and 5.69 deg. longitude in the Netherlands described in Su 20

et al. (2009). Unfortunately, we do not have observedFs and GPP for this period.

3.2 CCDAS simulations

Since the idealized tests may give a partial picture of the relationship betweenFs and

(14)

BGD

12, 707–749, 2015

E. N. Koffiet al.

Title Page

Tables Figures

◭ ◮

Back Close

Full Screen / Esc

Discussion

P

a

per

|

Discussion

P

a

per

|

Discussion

P

a

per

|

Discussion

P

a

per

|

relationship betweenFs and GPP is then investigated along with Vcmax and Cab. We

make simulations ofFs and GPP by using prior values ofVcmaxand their optimized

val-ues from Koffi_{et al. (2012). We also carry out simulations by using a constant value of}

Cab for all the 13 PFTs and a set ofCab values for each of them. We perform 4

exper-iments (i.e., S1 to S4), which are summarized in Table 2. The experexper-iments S1 and S3 5

use a constant value ofCab for all the 13 PFTs, while simulations S2 and S4 consider

C_ab to be PFT dependent (C_ab values are reported in Table 1). The experiments S1 and S2 consider the prior values ofVcmax, while S3 and S4 their optimized values. The

diff_{erences between S1 and S3 or between S2 and S4 give the sensitivity of} _F_s _and GPP toV_cmax. The diff_{erences between S1 and S2 or between S3 and S4 mainly give} 10

the sensitivity ofFstoCab.

The CCDAS simulates hourly Fs and GPP for one representative day in a month.

Since the computation of fluorescence is time consuming, we compute both Fs and

GPP only at 12 h local time, i.e., around the time of their peaks during a sunny day. For the simulatedFs, the computations are assigned to the 15th day of the month. We also

15

neglect the energy balance scheme in SCOPE which weakly aff_ects_F_s_.

4 Results

4.1 Idealized sensitivity tests using SCOPE

The results of these idealized sensitivity tests for the various LAI values are summa-rized in Figs. 2 and 3. For clarity, results from C3 plant are discussed. Then, some 20

conclusions are given for C4 plant.

4.1.1 Sensitivity ofFsand GPP to biochemistry parameters

As expected, both the fluorescence Fs and GPP increase with the increase of LAI

(Fig. 2). However, a weak sensitivity is found for LAI values greater than 4. As an illustration for the increase, for Vcmax=50 µmol m−

2

s−1_, _F

s values of 0.5 and

25

(15)

BGD

12, 707–749, 2015

E. N. Koffiet al.

Title Page

Tables Figures

◭ ◮

Back Close

Full Screen / Esc

Discussion

P

a

per

|

Discussion

P

a

per

|

Discussion

P

a

per

|

Discussion

P

a

per

|

1.25 W m−2_µm−1_sr−1 _{are found for LAI of 0.5 and 2, respectively (Fig. 2a). The} flu-orescence slightly increases with an increase ofVcmax. The sensitivity is relatively large

forVcmaxless than 70 µmol m− 2

s−1_{. Then,}_F

sremains almost constant forVcmaxhigher

than 125 µmol m−2_s−1_{(Fig. 2a). As an illustration, for LAI}₌_{2, the largest increase is of} only 50 % ofF_sforV_cmaxbetween 10 and 70 µmol m−2_s−1_{. Under the studied} configura-5

tionsFs increases withVcmaxwhen the GPP is controlled by the carboxylation enzyme

Rubisco, and remains almost constant when the electron transport rate is activated. GPP monotonically increases asV_cmaxincreases with large sensitivity for smallV_cmax

(less than 75 µmol m−2_s−1_{), then it becomes weakly sensitive for large values of}_V

cmax

(Fig. 2b). A moderate positive correlation is found betweenFs and GPP forVcmaxless

10

than 125 µmol m−2_s−1_{. Then, for larger}_V

cmax(i.e., 125 µmol m− 2

s−1_{), a very weak} neg-ative correlation betweenFs and GPP is obtained. The reason for this weak negative

correlation is thatFs slightly decreases for largeVcmax, while GPP even limited by the

carboxylation enzyme Rubisco still slightly increases (Fig. 2a and b). In fact, the value of irradiance at which the fluorescence at leaf levelΦ_Ft _{(Eq. 1) or}_F_s_{peaks increases} 15

with the increase ofVcmax. Thus, for the case presented in Fig. 2a with the short wave

radiationR_inof 500 W m−2_{, the peak of}_F

soccurs at aboutVcmax=200 µmol m− 2

s−1_. In the current version of the fluorescence model in SCOPE, the concentration of chlorophyllCab is set as a parameter and it is linked toFs through the transmittance

and reflectance of the leaves. Figure 2c portrays the variations ofF_s as a function of 20

Caband for various LAIs. For a given LAI,Fsincreases withCabwith large sensitivity for

Cab less than 20 µg cm− 2

. For largerCabvalues (i.e.,>50 µg cm− 2

),Fsremains almost

constant with a tendency to slightly decrease as C_ab increases. For a given C_ab, the variance inFsdue to the LAI can be significant.

Figure 2d displays GPP as a function of Cab (Fig. 2d). Except for small values of

25

C_ab (less than 5 µg cm−2_{), GPP is not sensitive to} _C

ab. The very weak sensitivity of

GPP toCabcomes from the impact of the chlorophyll content on the transmittance and

(16)

BGD

12, 707–749, 2015

E. N. Koffiet al.

Title Page

Tables Figures

◭ ◮

Back Close

Full Screen / Esc

Discussion

P

a

per

|

Discussion

P

a

per

|

Discussion

P

a

per

|

Discussion

P

a

per

|

4.1.2 Sensitivity ofFsand GPP to short wave radiation

For a given LAI, bothF_sand GPP increase with the top of canopy short wave radiation (Rin) (Fig. 2e and f). Thus, a strong positive linear correlation is obtained between Fs

andRin (Fig. 2e), while a non-linear (i.e., curvilinear) relationship is obtained between

GPP andR_in(Fig. 2f). For largeR_in, GPP increases with a slower rate indicating that the 5

photosynthesis is limited by the carboxylation enzyme Rubisco. For the selected values of LAI, large variance is found between Fs and Rin (Fig. 2f). We also investigate the

relationship between the simulated aPAR and both computedFsand GPP (not shown).

A very strong linear relationship betweenFsand aPAR is obtained. This relationship is

less sensitive to the LAI as it is for the relation betweenFsandRin(as shown in Fig. 2e).

10

GPP shows similar variations with aPAR as it does with the short wave radiation in Fig. 2f.

Finally, the sensitivities ofFsand GPP to aPAR for variousVcmaxare also investigated

(Fig. 3). A strong linear relationship betweenFsand aPAR is obtained with slopes which

are less sensitive to the values of Vcmax (Fig. 3a). Also, results clearly show that the

15

sensitivity ofFs toVcmax increases with the increase of aPAR, with almost no

sensitiv-ity for low values of aPAR (<250 W m−2_{). However, even with large values of aPAR,} the sensitivity of Fs toVcmax remains small. As expected, a curvilinear relationship is

found between GPP and aPAR with large variance in this relation for the selectedVcmax

(Fig. 3b). 20

The conclusions found from C3 plant relevant for the sensitivity of bothFs and GPP

to the input variables (Vcmax,Cab, andRin) are valid for C4 plant (not shown). However,

the amplitude of these sensitivities is slightly larger for C4 plant.

4.1.3 Simulations of in situ measurements

The time series of both simulatedFs and GPP for 16–20 June 2006 are presented in

25

Fig. 4. As expected, there is a strong correlation between aPAR and the short wave radiationRin (Fig. 4b), hence we discuss the results as a function of the observedRin.

(17)

BGD

12, 707–749, 2015

E. N. Koffiet al.

Title Page

Tables Figures

◭ ◮

Back Close

Full Screen / Esc

Discussion

P

a

per

|

Discussion

P

a

per

|

Discussion

P

a

per

|

Discussion

P

a

per

|

The temporal variations ofFsand GPP mainly follow that ofRin. Particularly, the

varia-tions ofFs mirror that ofRin, showing that the variance inFs due to the temperature is

low in this case study (Fig. 4a). At high irradiance GPP shows limitation by the carboxy-lation enzyme Rubisco, peaking early in the day whereasFsfollowsRin throughout the

day. The small variations in GPP at certain episodes can be explained by the temporal 5

variations of both the temperature and the vapour pressure (Fig. 4a). Note thatVcmax,

C_ab, and LAI are set constant during this period. Consequently, for this case study, the short wave radiation (hence aPAR) is the main driver of the relationship between Fs

and GPP. A curvilinear relation is obtained between GPP andFs. However, a relatively

strong linear correlation coeffi_{cient of 0.9 is derived. This suggests that} _F_s _{is a good} 10

constraint of GPP even if it does not directly constrainVcmax.

In summary, these idealized tests clearly show that the fluorescenceFsis more

sen-sitive to Cab, while GPP is more sensitive to Vcmax and both quantities are strongly

sensitive to the short wave radiation (or aPAR). However, GPP is limited by the car-boxylation enzyme Rubisco for large values of short wave radiation (or aPAR). Conse-15

quently, in this case the relationship betweenFs and GPP mainly driven by the short

wave radiation (or aPAR) is curvelinear. The part of the variance in this relationship due to the GPP can be explained byVcmax and environment conditions, while the variance

inFsis mainly due toCaband possibly to the geometrical parameters (i.e., solar zenith

angle and observation zenith angle) used in the retrieval ofFs.

20

4.2 CCDAS Simulations

To assess the relationship betweenFs and GPP at global scale, we perform the four

experiments described in Table 2. The results of these simulations are discussed along with the satellite-based Fs. We first analyze the correlations between the simulated

quantities and also the correlations between these simulations and the satellite based 25

Fs. Second, their mean spatial patterns are discussed and finally, the time series of

(18)

BGD

12, 707–749, 2015

E. N. Koffiet al.

Title Page

Tables Figures

◭ ◮

Back Close

Full Screen / Esc

Discussion

P

a

per

|

Discussion

P

a

per

|

Discussion

P

a

per

|

Discussion

P

a

per

|

4.2.1 Correlations betweenFsand GPP

For the discussion of the time series of modeled F_s and GPP at each CCDAS land pixel and the corresponding observedFs we analyze only pixels for which we have at

least one year satellite-basedFs data. Moreover, we consider only the time series of

these quantities for which the satellite-basedF_s data show consecutive values greater 5

than zero. Overall, the simulatedFs and GPP agree reasonably well with the

satellite-basedFs for most pixels. The seasonality of the satellite derivedFs is reasonably well

reproduced by both the simulatedFs and GPP as illustrated in Fig. 5. In accordance

with the idealized tests, the amplitudes of the satellite derivedFscan be better fitted by

appropriate values ofCab (Fig. 5a), while the simulated GPP is only weakly sensitive

10

to small Cab values as discussed in Sect. 4.1. As expected, the amplitudes of the

simulated GPP are strongly sensitive toVcmax(Fig. 5b).

We have computed the Pearson correlation coeffi_{cient between the time series of} satellite-basedFsand modelledFs and GPP at each pixel. For each pixel, we consider

only the pair of data for which the satellite-based Fs is greater than or equal to zero.

15

At most, 18 pairs of data are available for each pixel. We treat only pixels with at least 14 data points. Thus, a linear correlation is significant at least 10 % of level of significance for Pearson coeffi_{cient greater than 0.43. For about half of the 3462 land} pixels of CCDAS, the linear correlation coeffi_{cient between the satellite-based}_F_s _and either simulatedFs or GPP is small. For these latter pixels, we have analyzed the time

20

series of the satellite-basedFs(with their uncertainty) jointly with the simulatedFsand

GPP together with the aPAR as representative of the short wave radiation. For brevity sake, we only enumerate the diff_{erent cases without quantification since this does not} add anything valuable to our demonstration in the current study. We have cases for which:

25

– The peaks in simulated quantities (i.e., Fs and GPP) lag the satellite-based Fs

peak by at least one month. Other cases show opposite behavior.

(19)

BGD

12, 707–749, 2015

E. N. Koffiet al.

Title Page

Tables Figures

◭ ◮

Back Close

Full Screen / Esc

Discussion

P

a

per

|

Discussion

P

a

per

|

Discussion

P

a

per

|

Discussion

P

a

per

|

– The simulated Fs remain almost constant, while the satellite-based Fs show

a weak seasonality. Such cases predominantly occur in the tropics.

– The satellite-basedFs are larger (>2 W m− 2

µm−1_sr−1_{) than modeled} _F

s (around

1.2 W m−2_µm−1_sr−1_{). Such cases are mainly obtained in the tropics and for the} PFT 1 (i.e., tropical broadleaved evergreen tree).

5

– The simulated Fs are larger than satellite based Fs. Such cases are mainly

ob-tained from the PFT 9 (i.e., C3 grass).

– The satellite-based Fs show some unexpected peaks during period where they

are not expected and hence not modeled.

Second, we investigate the correlations between the simulated quantities (Fs, GPP, and

10

aPAR) at regional scales by using our best set up (i.e., experiment S4 in Table 2). We then assess the correlations between the simulated quantities (Fs, GPP, and aPAR)

and between simulated quantities and the satellite-basedFs. We select data at each

pixel such that the satellite-basedFsis greater or equal to zero and CCDAS land pixel

(i.e., the maximum fraction of coverage of the dominant PFT of the pixel) is greater 15

than zero. Data from June 2009 to end of 2010 are analyzed. We also give information about the dominant PFT of the pixels over the studied time period. To sample only over grid cells which are dominated by only one PFT, we consider only pixels for which the dominant PFT has a fraction of coverage greater than 50 %. Correlations are computed for the global and regional (Northern Hemisphere, tropics, and Southern Hemisphere) 20

regions and over the studied period. The results at global scale are shown in Fig. 6. A strong linear correlation is found between the computedFs and aPAR. This relation

is weakly sensitive to the PFTs (Fig. 6a). In contrast, the relationship between GPP and aPAR is PFT dependent (Fig. 6b). A good linear relationship between computed GPP and simulatedFs is obtained and again the slopes of this relationship are PFT

25

(20)

BGD

12, 707–749, 2015

E. N. Koffiet al.

Title Page

Tables Figures

◭ ◮

Back Close

Full Screen / Esc

Discussion

P

a

per

|

Discussion

P

a

per

|

Discussion

P

a

per

|

Discussion

P

a

per

|

The model SCOPE simulates quite well the observed Fs (Fig. 6d). However, large

observed Fs (>2 W m− 2

µm−1_sr−1_{) are not simulated. Such large observed} _F

s mainly

occur in the tropics. This result points out that short wave radiation used in the CCDAS simulations may be smaller than actual values. The contribution of chlorophyll content

Cab is low since the assigned value in tropics is already large (40 µg cm− 2

) and as 5

shown by the idealized tests, the simulated fluorescenceFs remains almost constant

forCabvalue larger or equal to 40 µg cm− 2

(Fig. 2c). The correlation coeffi_{cient between} modelled GPP and satellite-basedFs is 0.70 (Fig. 6e). This rises to 0.8 when we

ag-gregate both quantities to 4◦

×4◦ in agreement with Frankenberg et al. (2011). Finally,

as expected, a relatively good correlation is found between aPAR and satellite based 10

Fs(Fig. 6f).

Correlations are found to be larger between simulated quantities and satellite-derived

Fs in the Northern Hemisphere and moderate in the tropics and lower in the Southern

Hemisphere (not shown).

4.2.2 Mean spatial patterns ofFsand GPP 15

We compute the mean annual patterns of the satellite-basedFs and simulatedFs and

GPP for 2010. We discuss the simulated quantities by using the experiments S3 (i.e., optimizedVcmaxand constantCabfor all the 13 PFTs) and S4 (optimizedVcmaxandCab

PTF-specific) (See Table 2).

Figure 7 displays the annual mean observed and simulatedF_s,as well as simulated 20

GPP. Figure 7a shows the satellite basedFs. Figure 7b displays the modelledFsby

us-ing constantCabfor the 13 PFTs (experiment S3; Table 2), while Fig. 7c presents model

results ofF_sforC_abPTF-specific (experiment S4). Figure 7d exhibits the simulated GPP by using both Cab PFT-specific and optimized Vcmax (experiment S4). The model can

reasonably reproduce the mean spatial patterns of the satellite-basedFs with an

ap-25

propriate choice ofC_ab values for each of the 13 PFTs (Fig. 7a and c). The model with constantCab cannot reproduce the locations of maximum observedFs(Fig. 7a and b).

(21)

BGD

12, 707–749, 2015

E. N. Koffiet al.

Title Page

Tables Figures

◭ ◮

Back Close

Full Screen / Esc

Discussion

P

a

per

|

Discussion

P

a

per

|

Discussion

P

a

per

|

Discussion

P

a

per

|

Despite the good correlation, the computedFs with PFT-specificCab (Table 2) under-estimates the satellite-based data (Fig. 7a and c). Some of this mismatch corresponds to unlikely locations for satellite-derivedFs, e.g. central Australia.

A good agreement between the spatial patterns of GPP and satellite-based Fs is

found (Fig. 7a and d). Overall, we have a co-occurrence of hot spots of observedFs

5

and simulatedFsand GPP. Moreover, maximum simulatedFscoincides with maximum

APAR (not shown).

The small sensitivity of simulatedFstoVcmaxsuggests it may be difficult to use

obser-vations ofFs to constrain it. We can test this in a more realistic context by comparing

the diff_{erences between simulated} _F_s _{for prior and optimized values of} _V_cmax_{. If} dif-10

ferences are large compared to uncertainties in the observation thenFs observations

would allow constrainingVcmax. We compute the differences between simulated Fs by

using priorVcmax(experiment S2 in Table 2) and optimizedVcmax(experiment S4). Then,

we normalize these diff_{erences by the uncertainties in satellite based}_F_s_{. The derived} root mean square over year 2010 at pixel level can reach up to 67 % of the observed 15

uncertainties, but the global average is only 6 %. This suggests thatFsmeasurements

can only weakly constrainVcmaxwithin the current CCDAS.

4.2.3 Global and regional means ofFsand GPP

We compute the global and regional (i.e., Northern Hemisphere [30◦_–90◦_{N], Tropics} [30◦_S–30◦_{N] and Southern Hemisphere [90}◦_S–30◦_{S]) means at each month of the} 20

year and over June 2009 to December 2010 over land pixels. Results of both simulated

Fs and GPP from our best experimental set up (i.e., optimized Vcmax with Cab

PTF-specific; experiment S4 in Table 2) are discussed. The results show a reasonably good agreement between satellite-basedFsand both simulatedFsand GPP in terms of

sea-sonality (Fig. 8). However, on average, the simulated quantities peak one month earlier 25

than the peak of the satellite-basedFs(Fig. 8a). In the Northern Hemisphere,

satellite-basedFspeaks in July, while simulatedFsreaches its maximum in June (Fig. 8b). The

(22)

BGD

12, 707–749, 2015

E. N. Koffiet al.

Title Page

Tables Figures

◭ ◮

Back Close

Full Screen / Esc

Discussion

P

a

per

|

Discussion

P

a

per

|

Discussion

P

a

per

|

Discussion

P

a

per

|

the tropics, there is no significant seasonality in the satellite-based Fs, which is also

reproduced by the model (Fig. 9c). In the Southern Hemisphere, the satellite-basedFs

peaks in January, while modeled peaks in December (Fig. 8d). This weak seasonality shift in the CCDAS simulations is driven by the visible radiation at the top of the canopy (or aPAR) and LAI.

5

Quantitatively, the mean values of the simulated Fs are slightly smaller than that

of satellite-based (about 93 %) in the North hemisphere and the tropics. Since the above-mentioned regions dominated the amplitude ofFs, a good agreement between

simulated and satellite-basedFs is consequently found at global scale. The simulated

F_sin the Southern Hemisphere is about 1.47 times the value of satellite-basedF_s. The 10

main diff_{erences occur in Australia where the relatively large values of modeled}_F_s _are not shown in the satellite-basedFsdata (See Fig. 7a and c).

The zonal averages over the CCDAS land pixels of the satellite-based Fs and the

simulated quantities (Fs and GPP) are shown in Fig. 9. A good agreement is found

between the latitudinal variations of the satellite-basedFsand the simulatedFsby using

15

theCabPFT-specific (Fig. 9). Also, a good agreement is obtained between the

satellite-basedFs and the GPP (Fig. 9). All three quantities show maxima in the tropics and

around 45◦_{N. Simulated}_F

svalues are smaller than the satellite-basedFsin the tropics.

Between−15 and−45◦, the differences are mainly due to C4 grass for which both the

model’sVcmax andCab are apparently small. Around−35◦ latitude, the differences are

20

mainly due to the fact that the model simulates a largeFssignal over Australia, while the

satellite-based Fs shows only a small Fs signal. This discrepancy might be explained

by the uncertainty in the LAIs set to the evergreen shrub in the CCDAS in this area. Apparently, the LAIs in the CCDAS seem larger than expected values that give satellite basedFsmeasurements.

25

In summary, the agreement between simulated and observedFsis better as we move

to larger and larger scales.

(23)

BGD

12, 707–749, 2015

E. N. Koffiet al.

Title Page

Tables Figures

◭ ◮

Back Close

Full Screen / Esc

Discussion

P

a

per

|

Discussion

P

a

per

|

Discussion

P

a

per

|

Discussion

P

a

per

|

5 Discussions and concluding remarks

The first global maps ofFs retrieved from GOSAT measurements show promise in

es-timating the terrestrial gross photosynthetic uptake flux of CO2 (GPP) (Frankenberg

et al., 2011; Joiner et al., 2011). We have investigated the usefulness of these data in constraining GPP in the framework of CCDAS. We have augmented CCDAS with 5

SCOPE, which allows the calculation of GPP and Fs at leaf and canopy levels. In

CCDAS, the relationship between Fs and GPP is mediated by process parameters,

principally the maximum carboxylation capacity (Vcmax). Parameters not currently

in-cluded in CCDAS such as the chlorophyll content (Cab) of the leaves also affects the

observed fluorescence and so constitutes a nuisance variable in an assimilation ofFs

10

into CCDAS. We first calculate the sensitivity ofFsand GPP in the standalone SCOPE

model to a series of parameters, inputs or nuisance variables.Fs and GPP both

re-spond strongly to incoming radiation suggesting that, insofar as this input is uncertain,

Fs can provide a useful constraint. This uncertainty is currently not considered in the

CCDAS under study. 15

The relationship betweenVcmaxandFs is more complicated and weaker suggesting

that the conventional approach of using model parameters to mediate information from

Fs to GPP is unlikely to work. Cab also controls Fs while it has little impact on the

desired GPP making it a classical nuisance variable. Any model seeking to use Fs

should therefore account for chlorophyll concentration. 20

The simulations of CCDAS confirm the results from the idealized tests. Thus, the relationship between the simulated GPP and computedF_s is again found to be mainly controlled by the short wave radiation or aPAR. The analyses also show that a robust linear relationship betweenFsand GPP can be inferred for each PFT. This result is in

agreement with the finding of Guanter et al. (2012). 25

We compared observedFswith simulatedFsand GPP. The analyses showed a need

to select meaningful values for the chlorophyll content Cab for each of the 13 PFTs

(24)

BGD

12, 707–749, 2015

E. N. Koffiet al.

Title Page

Tables Figures

◭ ◮

Back Close

Full Screen / Esc

Discussion

P

a

per

|

Discussion

P

a

per

|

Discussion

P

a

per

|

Discussion

P

a

per

|

reproduction of the satellite-basedFs, with good co-location of the hot spots. Timing

of large-scale means is also good but this breaks down at pixel level. The global and regional as well as the zonal averages of the simulated quantities (Fs and GPP) are

in good agreement with the satellite-based Fs. On average, the peaks in simulated

Fs and GPP lag by one month the peaks in satellite-derived Fs in both southern and

5

Northern Hemispheres. The simulated quantities are found to be better correlated to the satellite based F_s when integrating the data at regional scales. More particularly, we found a significant linear correlation between simulated GPP and observedFs, but

a large scatter within the data is obtained. Such a variance can be attributed partly to the type of vegetation.

10

The study suggests some prospects for the use of satellite-based Fs to constrain

GPP. While we found a good correlation between the global and regional and zonal av-erages of simulated quantities and satellite-basedFs, we do not find a common process

parameter that propagates the information from the fluorescence to the GPP. Indeed, the relationship between GPP and satellite basedFsis mainly driven by the short wave

15

radiation or aPAR. Consequently, the mechanistic formulations of both Fs and GPP

under study do not allow us to constrain GPP throughVcmax.

Recent investigations by Zhang et al. (2014) show a very strong sensitivity ofFs to

Vcmax at in situ level using SCOPE version 1.52. Zhang et al. (2014) found about 4

times our sensitivity of Fs to Vcmax in the range of 20–200 µmol m− 2

s−1 _{as shown in} 20

our Figs. 2 and 3. We have modified our experiments to bring them closer to those of Zhang et al. (2014). First, Zhang et al. (2014) calculate Fs at 740 vs. 755 nm in this

study. Second Zhang et al. (2014) average their calculations from 09:00–12:00 local time, while we sample at 12:00. Results show that:

– The sensitivity of Fs to Vcmax is slightly larger at 740 nm than 755 nm and the

25

diff_{erence increases with aPAR. However, as an example, for a relatively large} aPAR (1400 W m−2_),_F

sat 740 nm is only 25 % higher thanFs at 755 nm. – The averaging period makes little diff_{erence to the sensitivity.}

(25)

BGD

12, 707–749, 2015

E. N. Koffiet al.

Title Page

Tables Figures

◭ ◮

Back Close

Full Screen / Esc

Discussion

P

a

per

|

Discussion

P

a

per

|

Discussion

P

a

per

|

Discussion

P

a

per

|

– Optimal choices of temperature and LAI produce a sensitivity about 2/3 that shown in Zhang et al. (2014). We would expect to reproduce their results so these diff_{erences remain under investigation.}

On the other hand, the results clearly show the good correlation between aPAR and both the fluorescence Fs and GPP, which support previous investigations. This both

5

points to a simpler application ofFs in constraining GPP and a problem with the

fore-going study. aPAR is an external forcing for BETHY which is taken to be well-known. Errors in forcing (like other nonparametric errors) are added to the observational error in CCDAS (Rayner et al., 2005), but the observations are unable to improve estimates of forcing. The parametric studies above hence miss a potential role of theFs

measure-10

ments in constraining GPP even if they cannot constrain process parameters.

Monteith (1972) proposed an empirical linear relation between GPP and aPAR which has been widely used by the satellite community to derive the GPP. The slope of this relationship is the effi_{ciency (}_ε_p_{) with which the absorbed radiation is converted to fixed} carbon.ε_pvaries with physiological stress. We have seen a strong linear relationship 15

between the fluorescenceFs and aPAR. Thus, the GPP is directly linked toFs by the

ratioεp/εf. Such an approach is described in a recent report of Berry et al. (2013). This approach would be easier to implement. It could be combined with other pertinent data for GPP (e.g., CO2 or Carbonyl sulfide (COS) concentration) within a simplified

CCDAS. This approach will be applied in a future study. 20

This study also shows a very weak sensitivity of GPP to the chlorophyll content (Cab)

present only for smallCab. This probably does not reflect reality. In the current version

the SCOPE model,Cab and Vcmaxare independent parameters, but in reality they are

correlated. In fact,Cabis related to the nitrogen content of the leaf which itself is linked

toVcmax(e.g., Kattge et al., 2009; Houborg et al., 2013). In addition, the nitrogen content

25

(26)

BGD

12, 707–749, 2015

E. N. Koffiet al.

Title Page

Tables Figures

◭ ◮

Back Close

Full Screen / Esc

Discussion

P

a

per

|

Discussion

P

a

per

|

Discussion

P

a

per

|

Discussion

P

a

per

|

6 Conclusions

We have investigated the usefulness of satellite derived fluorescence data to constrain GPP within CCDAS. We have coupled the SCOPE model to CCDAS to allow computing both fluorescenceFs and GPP. We have assessed the sensitivity of both Fs and GPP

to the environmental conditions at the interface of the canopy (short wave radiation 5

and meteorological variables) and the biophysical parameters (VcmaxandCab) by using

idealized and CCDAS simulations. Our results show:

– As expected, GPP is strongly sensitive toVcmax, whileFsis more sensitive to Cab

and only weakly sensitive toVcmax.

– The relationship between simulatedFs and GPP is mainly driven by aPAR. The

10

variance in this relationship is mostly explained by the Vcmaxand the chlorophyll

content. This highlights the need for better treatment of chlorophyll content in biosphere models.

– The global and regional means as well as the zonal averages of both simulated Fs and GPP are in good agreement with the satellite-basedFs.

15

– The seasonality of the satellite-basedFsis quite well reproduced by the simulated

Fsand GPP. However, the peaks of the simulated quantities lag by one month that

of the satellite-basedFsin the Northern and Southern Hemispheres.

– A good agreement is found between the simulated Fs and computed GPP. The

relationship is PFT dependent. 20

– A good agreement is found between the satellite-based F_s and the simulated quantities (Fs and GPP).

The study shows that the models of GPP andFsin the CCDAS built around SCOPE do

not allow us to propagate observations ofFsthrough constraint ofVcmaxto improve

es-timates of GPP. For this version of CCDAS, this study would rather recommend the use 25